# Claude Sonnet 5 vs gpt-oss-20b

> Claude Sonnet 5 is the stronger model overall, scoring 54.5 to 32.1 on the Noometry Index. gpt-oss-20b costs 111× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-sonnet-5-vs-gpt-oss-20b
- Last updated: 2026-10-10
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. Claude Sonnet 5 scores higher in 9 categories and gpt-oss-20b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude Sonnet 5 leads 69.4 to 35.8.
- The biggest single-benchmark swing is LMCA: 50% for Claude Sonnet 5 and 14.5% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Claude Sonnet 5 accepts more context: 1M tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 54.5 | 32.1 |
| Rank | 26 | 249 |
| Context | 1M | 131K |
| Input $/M | $2 | $0.018 |
| Output $/M | $10 | $0.09 |
| Weights | Proprietary | Open |

## Coding

- Claude Sonnet 5: 55.5 (#20)
- gpt-oss-20b: 37.1 (#185)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| SciCode | 53.6% | 34.4% |
| WeirdML | 68.8% | 40.9% |
| LMArena Coding | 1483 | 1307 |
| ALE-Bench | 1,463 | 566.05 |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| CursorBench | 34.1% | — |
| LMArena WebDev | 1540 | — |
| GSO | 37.3% | — |

## Agentic & Tool Use

- Claude Sonnet 5: 42.7 (#16)
- gpt-oss-20b: 9.2 (#146)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| APEX-Agents | 54.5% | — |
| GBAEval | 65.3% | — |
| LMArena Search | 1194 | — |
| Vending-Bench 2 | 6,378 | — |

## Reasoning

- Claude Sonnet 5: 48.2 (#37)
- gpt-oss-20b: 18.8 (#251)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| CritPt | 16.9% | 1.4% |
| Chess Puzzles | 35% | 4% |
| LMArena Hard Prompts | 1461 | 1274 |
| DTBench | 92.5% | 68% |
| LMCA | 50% | 14.5% |
| Epoch Capabilities Index | 156.34 | 137.81 |
| SimpleBench | 60.6% | — |
| Kagi LLM Benchmark | — | 53.2% |
| NYT Connections (extended) | 75.1% | — |
| Mystery Game Puzzles | 35% | — |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 61.1 | — |

## Math

- Claude Sonnet 5: 66.1 (#21)
- gpt-oss-20b: 37.6 (#93)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 94.7% | 65.3% |
| LMArena Math | 1467 | 1317 |
| FrontierMath (Tiers 1-3) | 65.6% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 77% | — |
| Omni-MATH | — | 56.5% |

## Knowledge

- Claude Sonnet 5: 55.7 (#42)
- gpt-oss-20b: 34.6 (#186)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 90.5% | 60.8% |
| LMArena Expert | 1491 | 1258 |
| SimpleQA Verified | 33.7% | — |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |

## Multimodal

- Claude Sonnet 5: 42.4 (#26)
- gpt-oss-20b: —

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1275 | — |
| Blueprint-Bench 2 | 24.9% | — |
| LMArena Document | 1466 | — |

## Multilingual

- Claude Sonnet 5: 54.0 (#55)
- gpt-oss-20b: 42.4 (#192)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1431 | 1269 |
| LMArena Chinese | 1475 | 1311 |
| LMArena German | 1448 | 1249 |
| LMArena Japanese | 1422 | 1245 |
| LMArena Korean | 1411 | 1239 |
| LMArena Russian | 1452 | 1279 |
| LMArena Spanish | 1436 | 1267 |
| LMArena French | 1457 | — |

## Instruction Following

- Claude Sonnet 5: 76.6 (#39)
- gpt-oss-20b: 62.0 (#235)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1453 | 1236 |
| IFEval | — | 73.2% |

## Long Context

- Claude Sonnet 5: 44.8 (#53)
- gpt-oss-20b: 37.9 (#203)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1464 | 1251 |

## Writing & Preference

- Claude Sonnet 5: 69.4 (#24)
- gpt-oss-20b: 35.8 (#260)

| Benchmark | Claude Sonnet 5 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1443 | 1288 |
| LMArena Creative Writing | 1416 | 1201 |
| EQ-Bench Creative Writing | 1794 | 666 |
| LMArena Multi-Turn | 1454 | 1268 |
| WildBench | — | 73.7% |
| EQ-Bench 4 | 1236 | — |

## FAQ

### Is Claude Sonnet 5 better than gpt-oss-20b?

Claude Sonnet 5 is the stronger model overall, scoring 54.5 to 32.1 on the Noometry Index. gpt-oss-20b costs 111× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

### Which is cheaper, Claude Sonnet 5 or gpt-oss-20b?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Claude Sonnet 5 lists at $2 and $10.

### Is Claude Sonnet 5 or gpt-oss-20b better for coding?

Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 37.1 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 5 does, with 1M tokens against 131K.

### How many benchmarks do Claude Sonnet 5 and gpt-oss-20b share?

27 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and gpt-oss-20b has 34.